❓ What X

Graph engineering designs an AI agent's control flow as nodes, edges, and shared state, so loops become visible, pausable, resumable, and testable.

AI evals explained: how datasets, graders, and scores tell you whether a language model feature works, and why vibe checks and benchmarks fall short.

Lima explained: how it runs Linux VMs on your Mac, powers Colima and Rancher Desktop, and replaces Docker Desktop without the licensing bill.

Envoy explained: what the proxy behind Istio and modern Kubernetes gateways does, why it exists, and how its control plane model works.

Boolean logic explained: true, false, AND, OR, NOT, and the mental patterns a human needs before true or false can mean anything.

A knowledge graph stores data as entities and relationships, not rows. Learn what it is, how it works, and when it's useful.

A software ontology is a shared, explicit model of a domain. Understand what it is in Domain-Driven Design terms, why it exists, and how it works.

AT Protocol explained for software developers: identity, repos, relays, AppViews, and lexicons, plus the trade-offs that matter when you build on it.

The cone of uncertainty narrows as projects progress. The hourglass of uncertainty says it flips again at launch. Learn which one your project is on.

Conway's Law says software mirrors the organization that built it. Learn the mechanism, the inverse Conway maneuver, and why team shape is architecture.

What is Ollama and how does it run large language models locally on consumer hardware?

fzf is a fuzzy finder that turns file selection into a fast, interactive search. Learn how it speeds up opening files in Neovim, macOS apps, and shell workflows.

Claude Code is Anthropic's agentic coding tool that runs in your terminal, edits real files, and executes commands. Here's how it works and why it exists.

Zoxide is a smarter cd that learns your most-used directories and jumps to them with a short query. Learn how frecency ranking replaces path typing.

nWave is an agentic AI methodology that runs specialized agents through six waves, with human review between each, enforcing TDD and peer review.

Coding agents explained: how harnesses, tools, memory, and context management make LLMs useful for real software engineering work.

Amdahl's Law sets the speed limit for parallelism. Learn why doubling cores doesn't double speed, how the serial fraction dominates, and when to optimize differently.

Just-in-Time Catching Test Generation creates tests on-the-fly with large language models to catch bugs before they land. Learn why it exists and how it works.

An Open Source Program Office (OSPO) coordinates strategy, compliance, and contributions. Learn what an OSPO is, why it exists, and how it works.

OpenSSF is a cross-industry initiative improving open source software security. Understand why it exists, how it works, and its role in securing the software supply chain.

SBOM is a Software Bill of Materials listing all components in software. Understand why it exists, how it works, and its role in software supply chains.

Exponential backoff is a retry strategy that increases wait time between attempts. Learn why it exists, how it works, and when to combine it with jitter.

Agent Skills explained: why they exist, how progressive disclosure loads them on demand, and how they relate to MCP, fine-tuning, and agent plugins.

Inoreader explained: an RSS feed reader with automation rules and filters that organize content automatically, solving information overload at scale.